Two new genera of the bee tribe Lonchopriini , with additional taxonomic notes ( Apidae , Colletinae )

Our comprehension of the evolution and biogeography of the apid subfamily Colletinae has greatly advanced through the fundamental works of ALMEIDA & DANFORTH (2009) and ALMEIDA ET AL. (2012, 2018). These studies have shed light on the relationships within the subfamily and provided a solid basis upon which a revised higher-level classification could be established. One of the major novelties brought by the molecular evidence was the position of the Australian genus Paracolletes Smith. It came out as sister group of the diphaglossine bees, a clade restricted to the New World, and not together within the remaining genera traditionally allocated in the Paracolletini, or Paracolletinae in the traditional classification (e.g. MICHENER 2007). With removal of Paracolletes, this aggregate of genera was placed under the name “Neopasiphaeinae” by ALMEIDA ET AL. (2012), containing three Australian and two New World lineages. Another significant finding derived from the molecular studies was the discovery that the New World “Neopasiphaeinae” were composed

Our comprehension of the evolution and biogeography of the apid subfamily Colletinae has greatly advanced through the fundamental works of ALMEIDA & DANFORTH (2009) and ALMEIDA ET AL. (2012, 2018. These studies have shed light on the relationships within the subfamily and provided a solid basis upon which a revised higher-level classification could be established. One of the major novelties brought by the molecular evidence was the position of the Australian genus Paracolletes Smith. It came out as sister group of the diphaglossine bees, a clade restricted to the New World, and not together within the remaining genera traditionally allocated in the Paracolletini, or Paracolletinae in the traditional classification (e.g. MICHENER 2007). With removal of Paracolletes, this aggregate of genera was placed under the name "Neopasiphaeinae" by ALMEIDA ET AL. (2012), containing three Australian and two New World lineages.
Another significant finding derived from the molecular studies was the discovery that the New World "Neopasiphaeinae" were composed Copyright by the Author. This is an open-access article distributed under the terms of the Creative Commons Attribution License (type CC-BY), which permits unrestricted use, distribution and reproduction in any medium, provided the original article is properly cited.

Taxonomia (Taxonomy)
of two separate lineages, one representing a large clade, encompassing most of the known diversity, and having as sister-group the Australian elements belonging to this clade. The other lineage was represented by Lonchopria sensu lato (in the sense of MICHENER 1989), which came out in a more basal position in relation to the larger clade mentioned above (see details in ALMEIDA ET AL. 2018). This lineage can be named as Lonchopriini, based on a family-group name proposed by MOURE (1945), but under a quite distinct scope (see also ENGEL 2020a for use of this name under an alternative classification).
The present contribution proposes two new genera in Lonchopriini. The species in these genera are found in Argentina and Brazil. Taxonomic notes are also provided for additional taxa in the tribe.

MATERIAL AND METHODS
The studied specimens belong to the DZUP -Coleção Entomológica Pe. Jesus Santiago Moure, Department of Zoology, Universidade Federal do Paraná, Curitiba, Brazil, and to the ZMB -Museum für Naturkunde, Berlin, Germany. The general morphological terminology follows URBAN (1967), SILVEIRA ET AL. (2002), andMICHENER (2007). The genus descriptions follow the format and character numbering used by MICHENER (1989 punctate in male. 3. Facial fovea not recognizable or indicated only by broad, undefined area of slightly different texture than rest of face. 4. Vertex in frontal view straight in female and weakly convex in male, elevated well above upper ocular tangent. Anterior margin of median ocellus below upper ocular tangent. Occipital carina absent. 5. Mandible slender with small preapical tooth on upper margin, not modified in male. 6. Labrum short, about four times as wide as long. 7. Proboscis without unusual features, glossa with short lobes; labial palpus at least as long as one half as prementum; maxillary palpus with two last articles extending beyond apex of galea. 8. Antenna of male reaching tegula, middle flagellar segments about as long as broad. 9. Metapostnotum finely transversely wrinkled, dull due to fine microreticulation; base of propodeum sloping, about as long as metanotum, curving onto declivous surface, about one third as long as declivous surface in profile. 10. Fore basitarsus of female without comb of hairs on outer margin. 11. Inner hind tibial spur of female and male ciliate, with many fine and short teeth. 12. Femoral scopa loose, with sparse long hairs with few side branches, femoral corbicula closed basally by somewhat floccose, curved hairs arising on trochanter. Female hind tibia with loose scopa, leaving most of tibial surface exposed, hair ramifications arising toward apical half; hairs of inner surface of hind tibia simple, relatively long and not forming zone of short keirotrichia. 13. Hind basitarsus of female gently tapering toward extreme apex which is about two-thirds as wide as width near base. Hind basitarsus of female with outer surface flat, vestiture distinct from that of tibia, hairs about as long as those of inner surface, not obscuring surface. 13A. Tarsal claws simple. 14. Basitibial plate of female well defined, not hidden by hair, pointed, about one fourth as long as tibia. 15. Basitibial plate of male well defined on both anterior and posterior margins. 16. Wing vestiture dense; two submarginal cells; first abscissa of vein M meeting cu-a. Pterostigma slender, distinctly broader than prestigma; marginal cell 1.7 times as long as pterostigma; vein 2r-rs arising near middle of pterostigma; margin within marginal cell convex. Apex of marginal cell narrowly rounded, separated from wing margin by a few vein widths. 17. Jugal lobe of hind wing slightly surpassing level of cu-a. 18. Metasomal sterna with abundant short hair and apical fringes of longer hair. 19. Pygidial plate of male represented by ill-defined elevated area that tapers toward apex of tergum 7 which is narrowly rounded; apex of female pygidial plate rounded. 20. Sternum 7 of male narrow medially, with two pairs of apical lobes, both rather narrow. 21. Sternum 8 of male with base rounded; apical process simple and pointed, not beveled like a pygidial plate. 22. Male genitalia with somewhat distinct gonostylus. Volsella with distinct denticles. Penis valves relatively large and elaborate, not strongly downcurved.

Etymology
The genus is named in honor of professor Lucio Antonio de Oliveira Campos, from the Universidade Federal de Viçosa, in recognition of his multifaceted contributions to the advancement of Brazilian apidology. http://dx.doi.org/10.5380/abp.v50i1-4.84026

Included species
Camposapis larejae (Compagnucci & Roig-Alsina, 2008) comb. nov., from Argentina, and an undescribed species from southern Brazil. The undescribed species was included in the study of ALMEIDA ET AL. (2018) and identified both as Bicolletes aff. larejae and Lonchopria aff. larejae.

Remarks
Within the Lonchopriini, Camposapis stands out for the female scopa composed of loose plumose setae, leaving most of the integument surface visible, forewing with two submarginal cells, front basitarsus of female without comb of hairs on outer margin, inner spur of hind tibia ciliate, and claws simple. In ALMEIDA ET AL. (2018)  Description 1. Integument mostly black, terga uniformly covered by decumbent pubescence and lacking metallic reflexes. Length 7-10 mm. 2. Face moderately convex, supraclypeal area not much elevated above clypeus but distinctly elevated above frons. Inner orbits converging below. Malar space very short. Clypeus with depressed, medial closely punctate area, lateral and distal to which are more shining convex areas. 3. Facial fovea indicated only by broad, undefined area of slightly different texture than rest of face. 4. Vertex weakly convex in frontal view, elevated well above upper ocular tangent. Anterior margin of medial ocellus below upper ocular tangent. Occipital carina absent. 5. Mandible of female slightly expanded apically, with small preapical tooth on upper margin, apex in male broadened by expansion of lower margin. 6. Labrum less than twice as broad as long. 7. Proboscis with glossa moderately elongate and distinctly bifid; labial palpus about one-quarter as long as prementum; maxillary palpus barely extending beyond apex of galea. 8. Antenna of male reaching little beyond tegula, middle flagellar segments 1.5 times as long as broad. 9. Metapostnotum mostly smooth and shining; base of propodeum sloping, about as long as metanotum, curving onto declivous surface, about one third as long as declivous surface in profile. 10. Front basitarsus of female with comb of hairs on outer margin. 11. Inner hind tibial spur of female pectinate with five very long teeth, their bases about as close as they can be so that they diverge from a sometimes somewhat thickened part of spur; of male ciliate with many fine and somewhat elongate teeth. 12. Femoral scopa of abundant, long hairs with numerous fine side branches, femoral corbicula closed basally by similar long, curved, plumose floccose hairs arising on trochanter. Female hind tibia with hairs of lower surface curled outward and upward, along with hairs of outer surface forming dense scopa of long, branched hairs entirely obscuring tibial surface; hairs of inner surface of hind tibia simple, moderately long, not forming zone of short keirotrichia. 13. Hind basitarsus of female tapering toward extreme apex which is only about half as wide as width near base. Hind (also mid) basitarsus of female with outer surface slightly longitudinally concave below upper margin, vestiture entirely different from that of tibia, hairs slightly shorter than those of inner surface, not obscuring surface. 13A. Tarsal claws bifid, inner ramus very short in female. 14. Basitibial plate of female well defined, hidden by hair, apex narrow to broadly rounded, about one-third as long as tibia or slightly less. 15. Basitibial plate of male well defined on both anterior and posterior margins. 16. Wing vestiture very sparse on membrane of closed cells, much denser on wing apex; three submarginal cells; first abscissa of vein M distal to cua. Pterostigma slender, distinctly broader than prestigma; marginal cell about 2.3 times as long as pterostigma; vein 2r-rs arising near middle of stigma, margin within marginal cell convex, somewhat angulate. Apex of marginal cell narrowly rounded, separated from wing margin by a few vein widths. 17. Jugal lobe of hind wing attaining level of cu-a. 18. Metasomal sterna with abundant short hair and apical fringes of longer hair. 19. Pygidial plate of male represented by ill-defined elevated area that tapers toward apex of tergum 7 which is emarginate; apex of female pygidial plate emarginate. 20. Sternum 7 of male narrow medially, with two pairs of apical lobes, basal pair broader. 21. S8 of male with base rounded; apical process simple and relatively broad, not beveled http://dx.doi.org/10.5380/abp.v50i1-4.84026 like a pygidial plate. 22. Male genitalia with somewhat distinct gonostylus. Volsella without denticles. Penis valves relatively large and elaborate, not strongly downcurved.

Etymology
The genus is named in honor of professor Fernando Amaral da Silveira, from the Universidade Federal de Minas Gerais, a cherished colleague and friend, whose scientific contributions have significantly impacted our current understanding of bee systematics.

Included species
Silveirapis robertsi (Michener, 1989) comb. nov., from Argentina, and perhaps an additional undescribed species, also from Argentina. Within the Lonchopriini, Silveirapis can be differentiated by the following combination of characters: (1) Clypeus with depressed, medial closely punctate area, lateral and distal to which are more shining convex areas; (2) Apex of male mandible broadened by expansion of lower margin; (3) Proboscis with glossa moderately elongate and distinctly bifid; (4) Inner hind tibial spur of female pectinate with five very long teeth; (5) Pubescence of outer surface of hind tibia forming dense scopa of long, branched hairs entirely obscuring tibial surface; (6) Outer surface of mid and hind basitarsus of female slightly longitudinally concave below upper margin; (7) Jugal lobe of hind wing attaining level of cu-a; (8) Terga uniformly covered by decumbent pubescence, tergal integument lacking metallic reflexes; (9) Apex of female pygidial plate and of male tergum 7 emarginate. Illustration of the male genitalia and associated sterna of S. robertsi can be found in MICHENER (1989).

Remarks
This taxon was chosen by COCKERELL (1914: 328) as the type species of Friese's Biglossa. Its identity was correctly interpreted by Moure (1948), who modified the classification of this group and segregated most species in the new genus Biglossidia, leaving only the type species in Biglossa. Unfortunately, MICHENER (1989) promoted a very conservative classification and lumped Biglossidia and Aeganopria under Biglossa and placed the latter as a subgenus of Lonchopria. The relationships within the Lonchopriini are still poorly known, but it is possible that Biglossa is more closely related to the clade formed by Silveirapis and Ctenosibyne. FRIESE (1906) stated in the original description that the taxon was based on 2 males and 12 females "von Salta 1200-2500 mtr; Steinbach leg.". In the ZMB collection, I located three females and two males that can be considered syntypes. One female and the two males bear a locality label as shown in Figs. 3C and 3F, while in the other two females the label is identical except for having "1200" instead of "2500". All specimens, except for one male, bear Friese's identification label (as shown in Figs. 3C and 3F). The female shown in Figs. 3A-B is here designated as lectotype. The specimen is in perfect conditions and clean. One male paralectotype is illustrated in Figs. 3D-F.   (Friese, 1906), lectotype female, presently designated (ZMB). A. Habitus, in lateral view. B. Habitus, in dorsal view. C. Specimen labels. Figures A and B at same scale.

Type material
Biglossidia chalybaea (Friese) (Figs. 4A-C) Biglossa chalybaea Friese, 1906: 378. Lectotype female, presently designated, Argentina: Salta (ZMB, examined). (1948) established the genus Biglossidia having Friese's Biglossa chalybaea as type species. He also kept Biglossa armata, described by Friese (1906) based on a single male also from Salta (Argentina), as a synonym of B. chalybaea. This synonymy has not been questioned, although it should be reevaluated considering the subtle differences observed in the forms occurring in the Argentinian Andes (see also MOURE 1948). Biglossidia is the largest genus in Lonchopriini and currently contains nine species.

MOURE
Type material FRIESE (1906) stated in the original description that the taxon was based on 3 females "von Salta, 2500 mtr., Steinbach leg.". In the ZMB collection, I located a single syntype, designated here as lectotype (Fig.  4). It is in perfect conditions. The labels of the specimen are shown in Fig. 4C.

Remarks
In addition to the species already placed in Biglossidia (see MOURE ET AL. 2007), the taxon described by FRIESE (1922) under the name Rhinetula chalybaea, from Mapiri (Bolivia), also belongs in this genus. It has been omitted from compilations (e.g., MICHENER 1989) and catalogs (e.g., MOURE ET AL. 1999, 2007 probably because of uncertainties about its identity. I was able to study the female holotype of this Friese's taxon and found out that it belongs in Biglossidia (Fig. 5). Due to the homonymy with the type species of the genus -Biglossidia chalybaea (Friese, 1906) -I propose here the name Biglossidia solanophila nom. nov. for Rhinetula chalybaea Friese, 1922. This species is quite distinct from B. chalybaea, what probably led Friese to describe it in a separate genus. Biglossidia solanophila is most similar to B. comforti (Gonzalez & Engel), from Colombia.
The new name is based on the species association with flowers of Solanum (Solanaceae) as source of pollen. I had the opportunity to collect two females of this species in Peru (new record: Cuzco, 19 km Type material Friese (1922) stated in the original description that he based the taxon on female "von Mapiri in Bolivia", indicating that he had a single specimen. The holotype is in good condition (Fig. 5), except for missing the 4th and 5th tarsomeres of the left mid leg. It is also somewhat dirt due to some fine white powder over the integument and pubescence. The specimen has an anomalous condition of the spurs in the left hind tibia. There is a single spur, which has an exceptionally enlarged base. Apparently, this resulted from fusion of the two spurs. In the right leg, there are two spurs and the inner one exhibits the usual morphology for the group. The labels of the specimen are shown in Fig. 5C. DISCUSSION We still lack a well-established classification at genus-level for the Lonchopriini. Under the concepts then prevailing for the colletine bees, MICHENER (1989) adopted a too conservative classification, with recognition of a single genus Lonchopria including five subgenera. MOURE ET AL. (1999) segregated these taxa, giving them genus-level status, and removed Aeganopria and Biglossidia under the synonymy with Biglossa.
In their system, a total of seven genera were recognized: Aeganopria, Biglossa, Biglossidia, Ctenosibyne, Lonchoprella Michener, Lonchopria and Porterapis Michener. This classification was followed in MOURE . More recently, ENGEL (2020a,b) retained a conservative approach to the classification of the Lonchopriini, recognizing only Lonchopria and Lonchoprella, and adding Lonchorhyncha Michener to the tribe, but without justifying why the latter genus should be included in this clade. Considering that Lonchorhyncha does not exhibit the few morphological characters that support the Lonchopriini, inclusive of Camposapis, it is not treated here as a member of this tribe.
Proposal here of two new genera expands our concepts for the Lonchopriini. While Silveirapis shares many features with the core Lonchopriini, Camposapis represents a considerably expansion of the tribal scope. At first glance, it looks quite an ordinary eulonchopriine bee and resembles the species placed in Bicolletes and Perditomorpha, es-http://dx.doi.org/10. 5380/abp.v50i1-4.84026 cially because of the two submarginal cells. Indeed, the type species, C. larejae, was associated with Perditomorpha by Compagnucci & Roig-Alsina (2008). An investigation of the morphological characters supporting the Lonchopriini, under the scope adopted here, is beyond the goals of this work and will be presented in a forthcoming contribution, which will also summarize the higher-level classification for the entire Colletinae. ACKNOWLEDGMENTS I thank Rodrigo Gonçalves, the editor of the Acta, for kindly handling this contribution. He and Kelli Ramos are thanked for reviewing the manuscript and for comments that helped improve it. I also thank Frank Koch and Michael Ohl for their kind help and support during my stay in the Museum für Naturkunde, Berlin. Financial support has been provided by Conselho Nacional de Desenvolvimento Científico e Tecnológico -CNPq (grants 205250/2014-9 and 309641/2016-0).

SUMMARY
Two new genera are proposed in Lonchopriini, a tribe of the neopasiphaeine line in the Colletinae. The new genus Camposapis (typespecies: Leioproctus larejae Compagnucci & Roig-Alsina, 2008) is proposed to accommodate a distinct lineage from Argentina and Brazil. The new genus Silveirapis (type-species: Lonchopria robertsi Michener, 1989) is proposed for a lineage previously placed in Biglossa sensu Michener, but which came out as sister-group of Ctenosibyne Moure based on molecular evidence. In addition, Biglossidia solanophila is proposed as a replacement name for Rhinetula chalybaea Friese, 1922 due to homonymy with Biglossidia chalybaea (Friese, 1906